CN108876842A - A kind of measurement method, system, equipment and the storage medium of sub-pixel edge angle - Google Patents

Abstract

This application discloses measurement method, system, equipment and the storage medium of a kind of sub-pixel edge angle, which includes：The original image for acquiring standardized element, establishes measurement model and adds the distribution parameter of the measured zone of the original image Plays part edge；The target image for acquiring part to be measured obtains the region to be measured of part edge to be measured in the target image；The region to be measured is aligned with the measured zone, extracts the marginal point in the region to be measured；Edge fitting is carried out to the region to be measured using RANSAC algorithm, obtains the sub-pixel edge angle of the part to be measured.The application carries out region alignment by the measurement model established, then carries out edge fitting by RANSAC algorithm, finally obtains sub-pixel edge angle, measurement model setting is simple, and calculation amount is few, improves the precision of measurement, and the robustness of system makes moderate progress, and measurement efficiency is high.

Description

Method, system, device and storage medium for measuring sub-pixel edge angle

technical field

The invention relates to the technical field of visual inspection, in particular to a method, system, device and storage medium for measuring sub-pixel edge angles.

Background technique

The dimension measurement of industrial products is an important link in the quality inspection of industrial products. In today's diversified and mass-produced assembly line production mode, manual inspection can no longer meet the requirements of product quality inspection in terms of cost, efficiency, and accuracy. Visual inspection is a new type of product inspection technology, which is based on image processing technology and has high flexibility and accuracy.

Visual inspection is to obtain image signals from image acquisition devices such as cameras, and then transmit them to the image processing system, process these information according to pixel distribution, brightness, color and other information to extract specific feature information, and finally control the device according to the processing results exercise. It is invaluable in detecting defects and more.

In general visual detection, after de-noising, binarization, threshold segmentation and other operations on the image, some edge detection algorithms are generally selected, such as the first-order Roberts Cross operator, Prewitt operator, Sobel operator, Kirsch Operators, etc., the second-order ones include Canny operators, Laplacian operators, etc. Most of the operators can detect the edge at the pixel level, but with the development of industrial technology, the traditional edge detection method can no longer meet the actual needs, so the sub-pixel level detection method emerges as the times require, it is usually for The image is binarized and analyzed, and then for the region that gradually changes excessively in the image, the sub-pixel position of the edge point is obtained by interpolation, fitting and other methods. Now the common methods include moment method, interpolation method and fitting method.

However, these methods are more sensitive to the noise of the image, or the calculation amount of the model is more complicated. For example, in general edge detection, after binary processing and threshold segmentation are performed on the image, in order to obtain a smoother edge line, dilation or erosion operations are usually performed on it, which results in low accuracy and robustness of part detection. problem of poor sex.

Therefore, how to solve the problems of low precision and complex model calculation in edge detection is a technical problem to be solved urgently by those skilled in the art.

Contents of the invention

In view of this, the object of the present invention is to provide a method, system, device and storage medium for measuring sub-pixel edge angles. The setting of the measurement model is simple, the amount of calculation can be reduced, the robustness of the system can be improved, and the accuracy of measurement can be improved. and efficiency. The specific plan is as follows:

A method for measuring sub-pixel edge angles, comprising:

collecting the original image of the standard part, establishing a measurement model and adding the distribution parameters of the measurement area on the edge of the standard part in the original image;

Collecting a target image of the part to be tested, and obtaining a region to be tested on the edge of the part to be tested in the target image;

aligning the area to be measured with the measurement area, and extracting edge points of the area to be measured;

The edge fitting of the area to be tested is carried out by using a random sampling consensus algorithm to obtain the sub-pixel edge angle of the part to be tested.

Preferably, in the above-mentioned sub-pixel edge angle measurement method provided in the embodiment of the present invention, the original image of the standard part is collected, the measurement model is established and the distribution parameters of the measurement area of the edge of the standard part in the original image are added, specifically including:

Collect original images of standard parts through industrial cameras;

Create a measurement model and add two intersecting measurement line objects to the measurement model;

determining the measurement area of the edge of the standard part in the original image according to the two measurement line objects;

performing grayscale and median filter processing on the original image to obtain an original grayscale image;

In the original grayscale image, the distribution parameters of the measurement area are acquired and added to the measurement model through threshold segmentation processing.

Preferably, in the above-mentioned sub-pixel edge angle measurement method provided in the embodiment of the present invention, the measurement area of the edge of the standard part in the original image is determined according to the two measurement line objects, specifically including:

Setting a measurement rectangle with a pre-defined range, the center of the measurement rectangle is the intersection of the two measurement line objects;

By stretching or contracting the two measuring line objects, the range of the measuring rectangle is adjusted to determine the measuring area of the edge of the standard part in the original image.

Preferably, in the above-mentioned sub-pixel edge angle measurement method provided in the embodiment of the present invention, the target image of the part to be tested is collected, and the area to be measured on the edge of the part to be measured in the target image is acquired, specifically including:

Collect the target image of the part to be tested through the industrial camera;

performing grayscale and median filter processing on the target image to obtain a grayscale image to be measured;

In the grayscale image to be tested, the region to be tested at the edge of the part to be tested in the target image is obtained through threshold segmentation processing.

Preferably, in the method for measuring the edge angle of the sub-pixel provided by the embodiment of the present invention, the region to be measured is within a predetermined range of the measurement rectangle.

Preferably, in the above-mentioned method for measuring the edge angle of the sub-pixel provided in the embodiment of the present invention, the region to be measured is aligned with the measurement region, and edge points of the region to be measured are extracted, which specifically includes:

using a vertex of said measurement area as a reference point;

performing shape or affine transformation alignment on the area to be measured and the measurement area according to the reference point, determining an edge position of the area to be measured, and extracting edge points of the area to be measured.

Preferably, in the method for measuring the sub-pixel edge angle provided in the embodiment of the present invention, the random sampling consistency algorithm is used to perform edge fitting on the region to be measured to obtain the sub-pixel edge angle of the part to be measured, Specifically include:

performing edge fitting on the area to be tested by using a random sampling consensus algorithm to obtain a sub-pixel edge profile of the part to be tested;

Two intersecting straight lines distributed on the sub-pixel edge contour are found to obtain the sub-pixel edge angle of the part to be tested.

The embodiment of the present invention also provides a sub-pixel edge angle measurement system, including:

The model building module is used to collect the original image of the standard part, establish a measurement model and add the distribution parameters of the measurement area on the edge of the standard part in the original image;

An area acquisition module, configured to acquire a target image of the part to be tested, and acquire the area to be tested on the edge of the part to be tested in the target image;

an area alignment module, configured to align the area to be measured with the measurement area, and extract edge points of the area to be measured;

The edge fitting module is used for performing edge fitting on the area to be tested by using a random sampling consensus algorithm to obtain the sub-pixel edge angle of the part to be tested.

The embodiment of the present invention also provides a virtual desktop display device, which is characterized by including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the above-mentioned sub A measure of the pixel edge angle.

The embodiment of the present invention also provides a computer-readable storage medium, which is characterized in that it is used to store a computer program; when the computer program is executed by a processor, the above-mentioned method for measuring the sub-pixel edge angle as provided by the embodiment of the present invention is realized .

A method, system, device, and storage medium for measuring sub-pixel edge angles provided by the present invention, the measurement method includes: collecting an original image of a standard part, establishing a measurement model and adding the measurement area of the edge of the standard part in the original image The distribution parameter of the part to be measured; collect the target image of the part to be tested, and obtain the area to be measured on the edge of the part to be measured in the target image; align the area to be tested with the measurement area, and extract the edge points of the area to be measured ; Using a random sampling consensus algorithm to perform edge fitting on the region to be tested to obtain a sub-pixel edge angle of the part to be tested.

The invention carries out area alignment through the established measurement model, and then performs edge fitting through the random sampling consistency algorithm, and finally obtains the sub-pixel edge angle. The measurement model setting is simple, the calculation amount is small, and the measurement accuracy is improved. The stickiness has been improved and the measurement efficiency is high.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention, and those skilled in the art can also obtain other drawings according to the provided drawings without creative work.

FIG. 1 is a flowchart of a method for measuring a sub-pixel edge angle provided by an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a sub-pixel edge angle measurement system provided by an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The present invention provides a method for measuring a sub-pixel edge angle, as shown in Figure 1, comprising the following steps:

S101. Collect the original image of the standard part, establish a measurement model and add the distribution parameters of the measurement area on the edge of the standard part in the original image;

In practical applications, the collected original image can be compared with the surrounding background of the standard part, and an image with high contrast is selected, so that the color interval can be easily divided in the subsequent threshold segmentation process, and the original edge of the standard part can be easily obtained; just The established measurement model is the data structure of an empty measurement model, and then the distribution parameters of the measurement area on the edge of the standard part in the original image are added to the data structure of the model; it should be noted that the calibration parameters of the camera, the pose of the part , the variables of the reference system, and related data such as the shape parameters of the parts to be measured and the distribution parameters of the area to be measured can be added in real time to the data structure of the measurement model as reference data to improve the calculation of the system speed;

S102. Collect a target image of the part to be tested, and obtain a region to be tested on the edge of the part to be tested in the target image;

Similarly, the collected target image can be to compare the part to be tested with the surrounding background, and select an image with high contrast, so that in the subsequent threshold segmentation process, the color interval is easy to be divided, and the original edge of the part to be tested is easy to obtain;

S103. Align the area to be measured with the measurement area, and extract edge points of the area to be measured;

In practical application, the extracted edge points of the area to be measured can be updated to the measurement model in real time;

S104. Using a random sampling consensus algorithm to perform edge fitting on the region to be tested to obtain a sub-pixel edge angle of the part to be tested.

In the method for measuring the above-mentioned sub-pixel edge angle provided by the embodiment of the present invention, first collect the original image of the standard part, establish a measurement model and add the distribution parameters of the measurement area of the edge of the standard part in the original image; then collect the part to be measured The target image of the target image is used to obtain the area to be measured on the edge of the part to be measured in the target image; then the area to be tested is aligned with the measurement area, and the edge points of the area to be measured are extracted; finally, the random sampling consistency is used The algorithm performs edge fitting on the area to be tested to obtain the sub-pixel edge angle of the part to be tested. In this way, the area alignment is carried out through the established measurement model, and then the edge fitting is carried out through the random sampling consensus algorithm, and finally the sub-pixel edge angle is obtained. The measurement model setting is simple, the calculation amount is small, the measurement accuracy is improved, and the system is robust. The performance has been improved and the measurement efficiency is high.

It should be noted that the sub-pixel edge angle can be understood as the angle between the edges of the region to be measured, which is an angle at the sub-pixel level. Here, the edges of the standard part and the part to be tested have a certain included angle, such as a triangular part, and the sub-pixel edge angle of the triangular part refers to the included angle between any two sides of the triangle. The above alignment is due to a certain position error in the placement of the parts to be measured. In order to accurately apply the measurement model, the target image needs to be straightened.

During specific implementation, in the above-mentioned sub-pixel edge angle measurement method provided by the embodiment of the present invention, step S101 collects the original image of the standard part, establishes a measurement model and adds the distribution parameters of the measurement area of the edge of the standard part in the original image , which may specifically include the following steps:

Step 1. Collect the original image of the standard part through the industrial camera;

In practical applications, industrial cameras are used to collect images, as long as the size of the original image can be obtained, the use of high-precision cameras can be avoided, and the cost can be reduced;

Step 2, establishing a measurement model, and adding two intersecting measurement line objects to the measurement model;

It should be noted that the two added measurement line objects are used to prepare for the subsequent delineation of the measurement rectangle range, which can avoid the original edge of the part being changed by the expansion and corrosion operation;

Step 3. Determine the measurement area of the edge of the standard part in the original image according to the two measurement line objects;

Specifically, when implementing step 3, firstly, a measurement rectangle with a pre-defined range can be set (as long as the measurement area is included, the specific measurement shape can be determined according to the actual situation), and the center of the measurement rectangle is two The intersection point of the measurement line object, so that the center of the measurement area is on the measurement line object; then by stretching or shrinking the two measurement line objects, adjusting the range of the measurement rectangle, the original image can be determined The measurement area of the edge of the standard part; the measurement area can be considered as a rectangular area containing a part of the edge image of the standard part;

It should be noted that, in order to reduce the amount of computation, the four sides of the measurement rectangle are kept parallel to the coordinate axes of the original image; that is, the two opposite sides of the measurement rectangle are parallel to the X-axis of the original image keep parallel; the other two opposite sides of the measurement rectangle keep parallel to the Y-axis of the original image;

Step 4, performing grayscale and median filter processing on the original image to obtain the original grayscale image;

Step 5. Obtain the distribution parameters of the measurement area through threshold segmentation processing in the original grayscale image and add them to the measurement model;

It should be noted that the measurement area of the part is described by threshold segmentation in the original grayscale image, and the specific threshold can be determined according to the actual situation; in the measurement model, some reference values can be changed, such as the calibration parameters of the camera, the pose, variables of the reference system, etc.

Further, during specific implementation, in the above-mentioned sub-pixel edge angle measurement method provided by the embodiment of the present invention, step S102 collects the target image of the part to be measured, and obtains the area to be measured on the edge of the part to be measured in the target image, Specifically, the following steps may be included:

Step 1. Collect the target image of the part to be tested through the industrial camera;

In practical applications, using industrial cameras to collect images can avoid the use of high-precision cameras and reduce costs;

Step 2, performing grayscale and median filter processing on the target image to obtain a grayscale image to be measured;

Step 3: Obtain the region to be tested at the edge of the part to be tested in the target image through threshold segmentation processing in the grayscale image to be tested;

It should be noted that the area to be measured can be within the pre-delimited range of the measurement rectangle, that is, the measurement rectangle with a pre-delimited range can be called to find the area on the edge of the part to be measured in the target image. The area to be tested; assuming that the target part is a triangular part, the area to be tested can be considered as a triangular area containing a part of the edge image of the triangular part; in addition, it should be noted that in order to improve the accuracy, the brightness of the light source should be be consistent.

Further, during specific implementation, in the above-mentioned sub-pixel edge angle measurement method provided by the embodiment of the present invention, step S103 aligns the area to be measured with the measurement area, and extracts edge points of the area to be measured , which may specifically include the following steps:

First, a vertex of the measurement area is used as a reference point;

Then, according to the reference point, the shape or affine transformation alignment of the area to be measured and the measurement area is performed to determine the edge position of the area to be measured, and the edge position of the area to be measured can be extracted from the edge position. edge point.

Further, in specific implementation, in the method for measuring the sub-pixel edge angle provided by the embodiment of the present invention, step S104 uses a random sampling consistency algorithm to perform edge fitting on the region to be measured, and obtains the The sub-pixel edge angle of the part may specifically include the following steps:

Firstly, using a random sampling consensus algorithm to perform edge fitting on the area to be tested to obtain the sub-pixel edge profile of the part to be tested;

Then, two intersecting straight lines distributed on the sub-pixel edge contour are found to obtain the sub-pixel edge angle of the part to be tested.

It should be noted that during specific implementation, in order to ensure the accuracy of the measurement, the poses of the parts to be measured should be kept consistent.

Based on the same inventive concept, an embodiment of the present invention also provides a measurement system for sub-pixel edge angles. Since the problem-solving principle of this system is similar to the aforementioned measurement method for sub-pixel edge angles, the implementation of the system can be found in sub-pixel The implementation of the method for measuring the pixel edge angle will not be repeated.

In specific implementation, the measurement system of the sub-pixel edge angle provided by the embodiment of the present invention, as shown in FIG. 2 , may specifically include:

The model building module 11 is used to collect the original image of the standard part, establish a measurement model and add the distribution parameters of the measurement area of the edge of the standard part in the original image;

The area acquisition module 12 is used to collect the target image of the part to be tested, and acquire the area to be tested on the edge of the part to be measured in the target image;

An area alignment module 13, configured to align the area to be measured with the measurement area, and extract edge points of the area to be measured;

The edge fitting module 14 is configured to use a random sampling consensus algorithm to perform edge fitting on the region to be tested to obtain a sub-pixel edge angle of the part to be tested.

In the above-mentioned sub-pixel edge angle measurement system provided by the embodiment of the present invention, the sub-pixel edge angle can be obtained through the interaction of the above four modules, the measurement model is simple to set, the amount of calculation is small, and the measurement accuracy is improved, and The robustness of the system is improved and the measurement efficiency is high.

Correspondingly, the embodiment of the present invention also discloses a virtual desktop display device, including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the sub-pixel edge angle disclosed in the foregoing embodiments is realized. Measurement methods.

For a more specific process of the above method, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not repeated here.

Furthermore, the present invention also discloses a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the method for measuring the sub-pixel edge angle disclosed above is implemented.

For a more specific process of the above method, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not repeated here.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same or similar parts of each embodiment can be referred to each other. As for the systems, devices, and storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and for relevant details, please refer to the description of the methods.

A method, system, device, and storage medium for measuring sub-pixel edge angles provided by the present invention, the measurement method includes: collecting an original image of a standard part, establishing a measurement model and adding the measurement area of the edge of the standard part in the original image The distribution parameter of the part to be measured; collect the target image of the part to be tested, and obtain the area to be measured on the edge of the part to be measured in the target image; align the area to be tested with the measurement area, and extract the edge points of the area to be measured ; Using a random sampling consensus algorithm to perform edge fitting on the region to be tested to obtain a sub-pixel edge angle of the part to be tested. The invention carries out area alignment through the established measurement model, and then performs edge fitting through the random sampling consistency algorithm, and finally obtains the sub-pixel edge angle. The measurement model setting is simple, the calculation amount is small, and the measurement accuracy is improved. The stickiness has been improved and the measurement efficiency is high.

Finally, it should also be noted that in this text, relational terms such as first and second etc. are only used to distinguish one entity or operation from another, and do not necessarily require or imply that these entities or operations, any such actual relationship or order exists. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The method for measuring the sub-pixel edge angle and its measurement system provided by the present invention have been described above in detail. In this paper, specific examples are used to illustrate the principle and implementation of the present invention. The descriptions of the above embodiments are only used to help understanding The method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation and scope of application. In summary, the content of this specification should not be construed as a limitation of the invention.

Claims (10)

1. a kind of measurement method of sub-pixel edge angle, which is characterized in that including：

The original image for acquiring standardized element, establishes measurement model and adds the measurement of the original image Plays part edge The distribution parameter in region；

The target image for acquiring part to be measured obtains the region to be measured of part edge to be measured in the target image；

The region to be measured is aligned with the measured zone, extracts the marginal point in the region to be measured；

Edge fitting is carried out to the region to be measured using RANSAC algorithm, obtains the sub-pix of the part to be measured Edge angle.

2. the measurement method of sub-pixel edge angle according to claim 1, which is characterized in that acquire the original of standardized element Beginning image establishes measurement model and adds the distribution parameter of the measured zone of the original image Plays part edge, specifically Including：

The original image of standardized element is acquired by industrial camera；

Measurement model is established, and the measurement line object that two intersect is added to the measurement model；

According to two measurement line objects, the measured zone of the original image Plays part edge is determined；

Gray processing and median filter process are carried out to the original image, obtain original-gray image；

It is handled in the original-gray image by Threshold segmentation, obtains the distribution parameter of the measured zone and be added to institute State measurement model.

3. the measurement method of sub-pixel edge angle according to claim 2, which is characterized in that according to two measurements Line object determines the measured zone of the original image Plays part edge, specifically includes：

The measurement rectangle to draw the line in advance is set, and the center of the measurement rectangle is the intersection point of two measurement line objects；

By stretching or shrinking two measurement line objects, the range of the measurement rectangle is adjusted, with the determination original graph As the measured zone of Plays part edge.

4. the measurement method of sub-pixel edge angle according to claim 3, which is characterized in that acquire the mesh of part to be measured Logo image obtains the region to be measured of part edge to be measured in the target image, specifically includes：

The target image of part to be measured is acquired by industrial camera；

Gray processing and median filter process are carried out to the target image, obtain gray level image to be measured；

It is handled in the gray level image to be measured by Threshold segmentation, obtains the to be measured of part edge to be measured in the target image Region.

5. the measurement method of sub-pixel edge angle according to claim 4, which is characterized in that the region to be measured is in institute It is interior to state drawing the line in advance for measurement rectangle.

6. the measurement method of sub-pixel edge angle according to claim 5, which is characterized in that by the region to be measured with The measured zone alignment, extracts the marginal point in the region to be measured, specifically includes：

One vertex of the measured zone is as a reference point；

According to the reference point, the region to be measured is subjected to shape with the measured zone or affine transformation is aligned, determines institute The marginal position for stating region to be measured extracts the marginal point in the region to be measured.

7. the measurement method of sub-pixel edge angle according to claim 6, which is characterized in that consistent using random sampling Property algorithm edge fitting is carried out to the region to be measured, obtain the sub-pixel edge angle of the part to be measured, specifically include：

Edge fitting is carried out to the region to be measured using RANSAC algorithm, obtains the sub-pix of the part to be measured Edge contour；

Two intersecting straight lines for being distributed in the sub-pixel edge profile are found out, obtain the sub-pixel edge angle of the part to be measured Degree.

8. a kind of measuring system of sub-pixel edge angle, which is characterized in that including：

Model building module, for acquiring the original image of standardized element, establishing measurement model and adding in the original image The distribution parameter of the measured zone at standardized element edge；

Region obtains module, for acquiring the target image of part to be measured, obtains part edge to be measured in the target image Region to be measured；

Region alignment module extracts the side in the region to be measured for the region to be measured to be aligned with the measured zone Edge point；

Edge fitting module obtains institute for carrying out edge fitting to the region to be measured using RANSAC algorithm State the sub-pixel edge angle of part to be measured.

9. a kind of virtual desktop shows equipment, which is characterized in that including processor and memory；Wherein, the processor executes Sub-pixel edge angle as described in any one of claim 1 to 7 is realized when the computer program saved in the memory Measurement method.

10. a kind of computer readable storage medium, which is characterized in that for storing computer program；The computer program quilt Processor realizes the measurement method of sub-pixel edge angle as described in any one of claim 1 to 7 when executing.